-
Notifications
You must be signed in to change notification settings - Fork 2.9k
Add kueue performance test job with TAS #36375
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
Add kueue performance test job with TAS #36375
Conversation
Add new performance tests with TopologyAwareScheduling: - pull-kueue-test-scheduling-perf-with-tas-main - pull-kueue-test-scheduling-perf-with-tas-release-0-15 - pull-kueue-test-scheduling-perf-with-tas-release-0-16
|
Welcome @ASverdlov! |
|
Hi @ASverdlov. Thanks for your PR. I'm waiting for a kubernetes member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: ASverdlov The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
| limits: | ||
| cpu: "6" | ||
| memory: "9Gi" | ||
| - name: pull-kueue-test-scheduling-perf-with-tas-main |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I assume we'll want to establish a proper baseline using the first couple of runs of this job, so I'm wondering whether we'll break our CI if the initial thresholds in kubernetes-sigs/kueue#8917 are misconfigured.
Should we temporarily make this job optional or maybe make the thresholds very lenient initially?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That is a valid concern, merging a PR which fails on every second build will not be good. The strategy we took with the baseline tests was:
- we run the CI a couple of times on Github to determine the mean and std dev, rather than from local machine
- we set the thresholds rather conservative, we want to detect bigger regressions, not 1% slow down due to a new feature. I assume the same level of tolerance should be used, like 20% or so from the mean.
MR adds new performance test jobs with TopologyAwareScheduling:
Don't merge before kubernetes-sigs/kueue#8917 is merged (and cherry-picked to 0.15, 0.16).